Pesticides in Selected Water-Supply Reservoirs and Finished Drinking Water, 1999-2000: Summary of Results from a Pilot Monitoring Program

Abstract

In 1999, a pilot monitoring program was initiated by the U.S. Geological Survey and the U.S. Environmental Protection Agency to provide information on pesticide concentrations in drinking water and to assist in the implementation of the Food Quality Protection Act of 1996. Drinking-water utilities that withdrew from reservoirs were sampled because reservoirs are vulnerable to pesticide contamination, integrate pesticide loading from their watersheds, and show smaller temporal variations than streams. Sampling frequencies were designed to measure long-term mean and short-term peak concentrations of pesticides in drinking water. Samples were collected quarterly throughout the year and at weekly or bi-weekly intervals following the primary pesticide-application periods. Water samples were collected from the raw-water intake and from the finished drinking-water tap prior to entering the distribution system. At some sites, samples were also collected at the reservoir outflow.

Twelve water-supply reservoirs were sampled, ranging in size from 120 to 92,600 acre-feet normal capacity, with contributing watersheds ranging from 3.3 to 784 square miles. The sites are in California, Indiana, Ohio, Oklahoma, Louisiana, Missouri, South Carolina, South Dakota, New York, North Carolina, Pennsylvania, and Texas. In 1999, drought conditions affected parts of the Eastern United States and California; therefore sampling was extended through 2000 at nine sites.

Analyses for pesticides in water samples were performed at the U.S. Geological Survey National Water Quality Laboratory in Denver, Colorado, using three analytical methods. Overall, these methods included 178 different pesticides and degradation products. Of these, 108 compounds were detected in at least one sample from a raw-water intake or finished-water tap. Many of these compounds, however, were detected in fewer than 5 percent of the samples and at concentrations very near the method reporting level. The observed detection frequencies for raw water are similar to those reported for 58 stream sites sampled from 1992-95 as part of the U. S. Geological Survey National Water-Quality Assessment Program. None of the pesticides were detected in finished water at a concentration greater than a National drinking-water criterion, such as a Maximum Contaminant Level or Health Advisory Level.

Widely used herbicides including atrazine, simazine, metolachlor, prometon, cyanazine, 2,4-D, tebuthiuron, and acetochlor were the most frequently detected pesticides in both intake and finished-water samples. Detection frequencies for these compounds ranged from 35.6 to 96.3 percent in raw-water samples and from 18.7 to 95.6 percent in finished-water samples. Four triazine-herbicide degradation products also were detected in more than 60 percent of all samples.

Diazinon and chlorpyrifos were the two most frequently detected insecticides in raw-water samples, with 35.3 percent and 5.3 percent detection frequencies, respectively. Neither of these insecticides were detected in finished water. Malaoxon, the principal oxidative transformation product of malathion, was detected in 11 finished-water samples and none of the raw-water samples. In three of the samples where malaoxon was detected in finished water, the corresponding intake sample showed measurable concentrations of malathion. Diazoxon, the principal oxidative transformation product of diazinon, was not included as an analyte on any of the three methods used in this study.